A Speaker-Dependent Approach to Single-Channel Joint Speech Separation and Acoustic Modeling Based on Deep Neural Networks for Robust Recognition of Multi-Talker Speech.
Yanhui TuJun DuChin-Hui LeePublished in: J. Signal Process. Syst. (2018)
Keyphrases
- speaker dependent
- speech recognition
- speaker independent
- single channel
- sound source
- speech enhancement
- speaker identification
- speech signal
- phoneme recognition
- neural network
- robust recognition
- speech recognizer
- hidden markov models
- speech recognition systems
- automatic speech recognition
- pattern recognition
- audio features
- audio visual
- multi channel
- acoustic features
- broadcast news
- speaker adaptation
- speaker recognition
- noisy environments
- independent component analysis
- gaussian mixture model
- speech synthesis
- text to speech
- high dimensional data
- machine learning